How to Choose a DSO Marketing Agency: A Multi-Location Buyer's Guide

I should say this up front: I co-founded a dental marketing company. So a guide about picking an agency, written by a guy who runs one, deserves some discount. I have tried to write the questions that are inconvenient for my own side of the table.

Almost every "how to choose a dental marketing agency" article, including the 10-point scorecard I wrote, is written for one practice. Cost per new patient. Call tracking. Who answers the phone. All of it matters, and none of it survives contact with twelve locations.

At scale the failure mode changes completely. A single practice knows when marketing is broken because the schedule is empty and the owner feels it that week. A group of twenty finds out eighteen months later, in a board deck, because the average looked fine the whole time.

---

The Number That Hides Everything

Run this arithmetic on your own group before your next vendor call.

Say you have twenty locations and a target of 60 new patients per location per month. Group target is 1,200. Your report comes back at 1,182. That is 98.5% of target, the slide is green, and everyone moves on to the next agenda item.

Now split it. Fourteen locations are running at 75 new patients. Six are running at 22. That is 1,050 plus 132, which is your 1,182.

Your fourteen strong offices are covering for six that are at roughly a third of target. Those six are bleeding chair time, the associates there are wondering why they are quiet, and the group-level report has never once shown it. The blended average did not just fail to catch the problem. It actively concealed it.

This is the single most useful thing to understand about multi-location marketing: aggregation is not a reporting style, it is a place where problems go to hide. Any agency that reports to you at the roll-up level by default is not measuring your business. They are measuring their own scorecard.

So the first question is not about strategy.

---

The First Question to Ask

"Pull up one location. Not the roll-up. Show me my worst-performing office and tell me why it is worst."

Watch what happens in the next ten seconds, because it tells you almost everything.

A vendor who works at scale already knows the answer. They will name the location, and the reason will be specific and boring: the office moved and the old address is still cited in forty places, or the GBP has been suspended since March, or that market has three DSO competitors and the paid CPCs are triple the group average, or the phones there are answered by voicemail after 4pm.

A vendor who does not work at scale will do one of three things. They will tell you they can put that together and send it over. They will open the group dashboard and start filtering in front of you. Or they will pivot to explaining why location-level data is noisy and the trend matters more.

That third answer is the most dangerous one, because it is partly true. Location-level data is noisier. It is also the only level at which you can actually do anything.

---

What Breaks at Scale That Nobody Asks About

Brand architecture is a one-way door

One domain for the group, or a separate site per location? Every DSO faces this and most treat it as a branding preference. It is not. It is the single most expensive technical decision in multi-location dental marketing, and reversing it costs you a year.

I will give you my own example of getting this wrong. I recommended splitting a two-location group onto separate domains. I had a clean rationale for it. Then I actually tested the split four different ways before we executed, and it lost every single time. I went back to the client and reversed my own recommendation.

The right answer depends on your market overlap, your existing authority, your acquisition pipeline, and whether the practices carry real local brand equity. Any agency that answers this in under a minute, in either direction, has a template rather than a method. Ask what they would need to look at first. If they cannot list four things, keep interviewing.

The Google Business Profile portfolio is the real asset

At one location, your GBP is a marketing channel. At forty, the portfolio is infrastructure, and the way it is held determines whether you can ever leave.

Ask specifically: is every listing inside a Google business group that the DSO owns, with the agency added as a manager? Or did the agency create the group and grant you access to it? Those two arrangements look identical on a screenshare and could not be more different when you switch vendors.

The difference is not a matter of opinion, it is in Google's own permission model. A business group manager can add, edit, and delete every profile in the group, but only a group owner can delete the group or hand it to someone else. And a primary owner cannot even remove themselves until they transfer that ownership to another user. So if your agency holds primary ownership of the group that contains forty locations, your exit is a favor they do you rather than a button you press.

Same question for your ad accounts, your analytics, and your call tracking numbers. Those last ones are what people always forget, and they walk out with your historical data attached.

I wrote up the contract clauses to catch here, and every one of them gets worse when it is multiplied by thirty locations.

Acquisition intake is a capability, not a favor

If you are growing by acquisition, you will hand your agency five practices at once, each with a different website platform, a different PMS, a mess of duplicate listings, and a previous agency that owns half the accounts.

Ask what their intake process is, how long a location takes from close to live, how many they can absorb in a month without the existing locations degrading, and what it costs. A vendor with real multi-location experience has a number for all four. A vendor without one will say they are flexible.

Then ask the reverse, which nobody asks: when you divest a location, what happens? Does it come off the agreement cleanly, does the buyer inherit the assets, and are you still paying for it next quarter?

Cannibalization between your own locations

Two of your offices are eleven minutes apart. They are bidding on the same keywords, ranking for the same "dentist near me" searches, and one of them is winning at the other's expense. Your group total looks flat and your agency reports both locations as "performing."

Ask whether they report on internal competition at all. Most do not, because it is a report where their own work looks bad.

---

The Questions, Compressed

Take these into the meeting as they are:

1. Show me my worst location and tell me why it is worst.
2. Which of my locations would you drop if we cut scope by 20%, and why that one?
3. Who owns the GBP business group, the ad accounts, the analytics, and the call tracking numbers?
4. How many new locations can you onboard in a month before existing locations degrade?
5. What is your process and price when I close on five practices at once?
6. What happens to the agreement and the assets when I divest a location?
7. How do you attribute a patient who searches the group brand and books at a different office than the one they searched?
8. Do you report on my locations competing with each other? Show me.
9. Who writes location-level content, and does anyone local review it before it publishes?
10. What breaks first when we double our location count?

Question ten is my favorite. Anyone who has genuinely scaled a group has a fast, specific, slightly weary answer, because they have watched it break. "Your onboarding queue" or "location-level content review" or "your call handling." Someone who has not will tell you nothing breaks because their process scales.

Nothing breaks is never the answer. Something always breaks first.

---

Red Flags Specific to Multi-Location

Every location page is one template with the city name swapped. I audit a lot of group sites and this is the most common thing I find. It is also easy to check yourself right now: open three of your location pages and read the body copy. If the only difference is the city and the phone number, you do not have thirty location pages. You have one page thirty times, and Google treats it accordingly. Sometimes the swap is not even complete and every page still names the flagship city.

Reporting that only exists at the group level. Covered above. It is the big one.

A master agreement with no location schedule. If the contract does not enumerate which locations are covered at what price, you cannot cleanly add, remove, or renegotiate any single one of them.

Per-location pricing with no volume logic. Location twenty-five costs the same as location two, and nobody can explain what extra work location twenty-five requires. Either there are real economies of scale and you should see them, or the work is genuinely linear and they should be able to say why.

No named owner for your account. At group scale, "the team will handle it" means requests route to whoever is free and nobody holds the history of why your Fresno office is structured differently.

---

Do Not Sign Twenty Locations

This is the part I would push hardest if you were paying me instead of reading me for free.

Whatever you are about to sign for the whole group, sign it for three locations first. Pick your best performer, your worst performer, and one in the middle. Ninety days. Define what success looks like per location before you start, in writing, with the actual numbers.

You learn three things you cannot learn from a pitch. Whether they can lift the worst one, which is the only real test of skill. Whether they can avoid breaking the best one, which is the only real test of discipline. And whether their reporting tells you the truth about the median one, which is the test of whether you can trust anything else they send you.

A vendor who is confident in their work will take that deal. A vendor who needs all twenty locations signed to make the economics work is telling you their economics are the point.

If you are weighing whether to build this internally instead, I compared in-house, agency, and fractional here.

---

Where I Land

The DSO marketing problems I see are almost never marketing problems. They are infrastructure problems wearing a marketing costume. Visibility gaps, reporting blind spots, brand inconsistency, all of it traces back to a decision made at four locations by someone who was not thinking about forty.

So evaluate vendors on infrastructure, not on campaign ideas. Anyone can bring campaign ideas. Very few can tell you what breaks when you double, because that answer only comes from having been there when it broke.

For the strategy side rather than the vendor-selection side, the rest of the multi-location cluster is here.

Full disclosure again, since I opened with it: I run Lasso MD, and we work with multi-location groups. I would rather you take these ten questions to four agencies including us and pick on the answers than take my word for any of it. If we are not the right fit for your group, the questions still work.

If your group or association wants this as a talk rather than a blog post, that is something I speak on.

Sources

• Own & manage business groups: Google Business Profile Help, the owner and manager capability table for business groups, including who can delete a group and the primary-owner transfer requirement.
• About business groups: Google Business Profile Help, what a business group is and how organizations use them to manage profiles for multiple locations at once.